Performance improvements

[reline]

Currently, the search for "house" takes a considerable amount of time to complete (longer than five seconds). Research improving the query speed.

Notably from what I can see, the two areas of impact are the subquery and the loop of queries (senses, glosses, pos).

SELECT Entry.id, Entry.is_common AS isCommon, Entry.kanji, Entry.reading
FROM Entry
WHERE Entry.id IN (
    SELECT sub.id
    FROM Entry AS sub
    LEFT JOIN Sense
        ON sub.id = Sense.entry_id
    LEFT JOIN Gloss
        ON Gloss.sense_id = Sense.id
    WHERE Gloss.value LIKE '%'||:keyword||'%'
);

val entries = database.entryQueries.selectEntriesByGloss(query, ::Mapper).executeAsList()
...
          entries.map { entry ->
                ...
                // here we query at least once for each entry found, then twice for each sense
                // "house" returns over 1000 entries, so that's at least 3000 queries
                val senses = database.senseQueries.selectSenses(entry.id).executeAsList().map { senseId ->
                    val pos = database.sensePosTagQueries.selectPosWhereSenseIdEquals(senseId).executeAsList()
                    val glosses = database.glossQueries.selectGlossWhereSenseIdEquals(senseId).executeAsList()
                    Sense(glosses, pos)
                }
                ...
          }